The PhenoImager HT was purchased through funding provided by the Radboud University Medical Center and Radboud Technology Center for Microscopy. CF is financially supported by a Dutch Cancer Society grant (10673) and ERC Adv grant ARTimmune (834618). JT is financially supported by an NWO Vidi grant (VI.Vidi.192.084). The authors would like to thank Eric van Dinther and Ankur Ankan for their assistance in creating workflows to store multiplex IHC data and Bengt Phung is thanked for instructions on how to implement multiplex IHC data in QuPath for ROI drawing.

Mapping the Tumor Immune Landscape: An In-Depth Multiplex IHC Analysis Protocol

<ol>
	<li>Dunn, G. P., Bruce, A. T., Ikeda, H., Old, L. J., Schreiber, R. D. <a target="_blank" href="http://www.ncbi.nlm.nih.gov/entrez/query.fcgi?db=PubMed&cmd=Search&doptcmdl=Citation&defaultField=Title+Word&term=Cancer+immunoediting:+from+immunosurveillance+to+tumor+escape.">Cancer immunoediting: from immunosurveillance to tumor escape.</a> Nature Immunology. 3 (11), 991-998 (2002).</li><li>van der Woude, L. L., Gorris, M. A. J., Halilovic, A., Figdor, C. G., de Vries, I. J. M. <a target="_blank" href="http://www.ncbi.nlm.nih.gov/entrez/query.fcgi?db=PubMed&cmd=Search&doptcmdl=Citation&defaultField=Title+Word&term=Migrating+into+the+tumor:+a+roadmap+for+T+cells.">Migrating into the tumor: a roadmap for T cells.</a> Trends in Cancer. 3 (11), 797-808 (2017).</li><li>Fridman, W. H., Pages, F., Sautes-Fridman, C., Galon, J. <a target="_blank" href="http://www.ncbi.nlm.nih.gov/entrez/query.fcgi?db=PubMed&cmd=Search&doptcmdl=Citation&defaultField=Title+Word&term=The+immune+contexture+in+human+tumours:+impact+on+clinical+outcome.">The immune contexture in human tumours: impact on clinical outcome.</a> Nature Reviews. Cancer. 12 (4), 298-306 (2012).</li><li>Fridman, W. H., et al. <a target="_blank" href="http://www.ncbi.nlm.nih.gov/entrez/query.fcgi?db=PubMed&cmd=Search&doptcmdl=Citation&defaultField=Title+Word&term=The+immune+microenvironment+of+human+tumors:+general+significance+and+clinical+impact.">The immune microenvironment of human tumors: general significance and clinical impact.</a> Cancer Microenvironment. 6 (2), 117-122 (2013).</li><li>Pages, F., et al. <a target="_blank" href="http://www.ncbi.nlm.nih.gov/entrez/query.fcgi?db=PubMed&cmd=Search&doptcmdl=Citation&defaultField=Title+Word&term=International+validation+of+the+consensus+Immunoscore+for+the+classification+of+colon+cancer:+a+prognostic+and+accuracy+study.">International validation of the consensus Immunoscore for the classification of colon cancer: a prognostic and accuracy study.</a> Lancet. 391 (10135), 2128-2139 (2018).</li><li>Angell, H. K., Bruni, D., Barrett, J. C., Herbst, R., Galon, J. <a target="_blank" href="http://www.ncbi.nlm.nih.gov/entrez/query.fcgi?db=PubMed&cmd=Search&doptcmdl=Citation&defaultField=Title+Word&term=The+Immunoscore:+colon+cancer+and+beyond.">The Immunoscore: colon cancer and beyond.</a> Clinical Cancer Research. 26 (2), 332-339 (2020).</li><li>Angell, H., Galon, J. <a target="_blank" href="http://www.ncbi.nlm.nih.gov/entrez/query.fcgi?db=PubMed&cmd=Search&doptcmdl=Citation&defaultField=Title+Word&term=From+the+immune+contexture+to+the+Immunoscore:+the+role+of+prognostic+and+predictive+immune+markers+in+cancer.">From the immune contexture to the Immunoscore: the role of prognostic and predictive immune markers in cancer.</a> Current Opinion in Immunology. 25 (2), 261-267 (2013).</li><li>Galon, J., et al. <a target="_blank" href="http://www.ncbi.nlm.nih.gov/entrez/query.fcgi?db=PubMed&cmd=Search&doptcmdl=Citation&defaultField=Title+Word&term=Cancer+classification+using+the+Immunoscore:+a+worldwide+task+force.">Cancer classification using the Immunoscore: a worldwide task force.</a> Journal of Translational Medicine. 10, 205 (2012).</li><li>Galon, J., et al. <a target="_blank" href="http://www.ncbi.nlm.nih.gov/entrez/query.fcgi?db=PubMed&cmd=Search&doptcmdl=Citation&defaultField=Title+Word&term=World-wide+Immunoscore+Task+Force:+meeting+report+from+the+34;Melanoma+Bridge&amp;#34,+Napoli,+November+30th-December+3rd,+2016.">World-wide Immunoscore Task Force: meeting report from the 34;Melanoma Bridge&amp;#34, Napoli, November 30th-December 3rd, 2016.</a> Journal of Translational Medicine. 15 (1), 212 (2017).</li><li>Tumeh, P. C., et al. <a target="_blank" href="http://www.ncbi.nlm.nih.gov/entrez/query.fcgi?db=PubMed&cmd=Search&doptcmdl=Citation&defaultField=Title+Word&term=PD-1+blockade+induces+responses+by+inhibiting+adaptive+immune+resistance.">PD-1 blockade induces responses by inhibiting adaptive immune resistance.</a> Nature. 515 (7528), 568-571 (2014).</li><li>Creemers, J. H. A., et al. <a target="_blank" href="http://www.ncbi.nlm.nih.gov/entrez/query.fcgi?db=PubMed&cmd=Search&doptcmdl=Citation&defaultField=Title+Word&term=A+tipping+point+in+cancer-immune+dynamics+leads+to+divergent+immunotherapy+responses+and+hampers+biomarker+discovery.">A tipping point in cancer-immune dynamics leads to divergent immunotherapy responses and hampers biomarker discovery.</a> Journal for Immunotherapy of Cancer. 9 (5), e002032 (2021).</li><li>Blank, C. U., Haanen, J. B., Ribas, A., Schumacher, T. N. <a target="_blank" href="http://www.ncbi.nlm.nih.gov/entrez/query.fcgi?db=PubMed&cmd=Search&doptcmdl=Citation&defaultField=Title+Word&term=CANCER+IMMUNOLOGY.+The+&amp;#34;cancer+immunogram&amp;#34;.">CANCER IMMUNOLOGY. The &amp;#34;cancer immunogram&amp;#34;.</a> Science. 352 (6286), 658-660 (2016).</li><li>Teruya-Feldstein, J. <a target="_blank" href="http://www.ncbi.nlm.nih.gov/entrez/query.fcgi?db=PubMed&cmd=Search&doptcmdl=Citation&defaultField=Title+Word&term=The+immunohistochemistry+laboratory+looking+at+molecules+and+preparing+for+tomorrow.">The immunohistochemistry laboratory looking at molecules and preparing for tomorrow.</a> Archives of Pathology &amp; Laboratory Medicine. 134 (11), 1659-1665 (2010).</li><li>Ramos-Vara, J. A., Miller, M. A., et al. <a target="_blank" href="http://www.ncbi.nlm.nih.gov/entrez/query.fcgi?db=PubMed&cmd=Search&doptcmdl=Citation&defaultField=Title+Word&term=When+tissue+antigens+and+antibodies+get+along:+revisiting+the+technical+aspects+of+immunohistochemistry--the+red,+brown,+and+blue+technique.">When tissue antigens and antibodies get along: revisiting the technical aspects of immunohistochemistry--the red, brown, and blue technique.</a> Veterinary Pathology. 51 (1), 42-87 (2014).</li><li>Hegde, P. S., Karanikas, V., Evers, S. <a target="_blank" href="http://www.ncbi.nlm.nih.gov/entrez/query.fcgi?db=PubMed&cmd=Search&doptcmdl=Citation&defaultField=Title+Word&term=The+where,+the+when,+and+the+how+of+immune+monitoring+for+cancer+immunotherapies+in+the+era+of+checkpoint+inhibition.">The where, the when, and the how of immune monitoring for cancer immunotherapies in the era of checkpoint inhibition.</a> Clinical Cancer Research. 22 (8), 1865-1874 (2016).</li><li>Parra, E. <a target="_blank" href="http://www.ncbi.nlm.nih.gov/entrez/query.fcgi?db=PubMed&cmd=Search&doptcmdl=Citation&defaultField=Title+Word&term=Novel+platforms+of+multiplexed+immunofluorescence+for+study+of+paraffin+tumor+tissues.">Novel platforms of multiplexed immunofluorescence for study of paraffin tumor tissues.</a> Journal of Cancer Treatment and Diagnosis. 2 (1), 43-53 (2018).</li><li>Gorris, M. A. J., et al. <a target="_blank" href="http://www.ncbi.nlm.nih.gov/entrez/query.fcgi?db=PubMed&cmd=Search&doptcmdl=Citation&defaultField=Title+Word&term=Eight-color+multiplex+immunohistochemistry+for+simultaneous+detection+of+multiple+immune+checkpoint+molecules+within+the+tumor+microenvironment.">Eight-color multiplex immunohistochemistry for simultaneous detection of multiple immune checkpoint molecules within the tumor microenvironment.</a> Journal of Immunology. 200 (1), 347-354 (2018).</li><li>Roelofsen, T., et al. <a target="_blank" href="http://www.ncbi.nlm.nih.gov/entrez/query.fcgi?db=PubMed&cmd=Search&doptcmdl=Citation&defaultField=Title+Word&term=Spontaneous+regression+of+ovarian+carcinoma+after+septic+peritonitis;+a+unique+case+report.">Spontaneous regression of ovarian carcinoma after septic peritonitis; a unique case report.</a> Frontiers in Oncology. 8, 562 (2018).</li><li>van den Brand, D., et al. <a target="_blank" href="http://www.ncbi.nlm.nih.gov/entrez/query.fcgi?db=PubMed&cmd=Search&doptcmdl=Citation&defaultField=Title+Word&term=Peptide-mediated+delivery+of+therapeutic+mRNA+in+ovarian+cancer.">Peptide-mediated delivery of therapeutic mRNA in ovarian cancer.</a> European Journal of Pharmaceutics and Biopharmaceutics. 141, 180-190 (2019).</li><li>van den Brand, D., et al. <a target="_blank" href="http://www.ncbi.nlm.nih.gov/entrez/query.fcgi?db=PubMed&cmd=Search&doptcmdl=Citation&defaultField=Title+Word&term=EpCAM-binding+DARPins+for+targeted+photodynamic+therapy+of+ovarian+cancer.">EpCAM-binding DARPins for targeted photodynamic therapy of ovarian cancer.</a> Cancers. 12 (7), 1762 (2020).</li><li>Di Blasio, S., et al. <a target="_blank" href="http://www.ncbi.nlm.nih.gov/entrez/query.fcgi?db=PubMed&cmd=Search&doptcmdl=Citation&defaultField=Title+Word&term=The+tumour+microenvironment+shapes+dendritic+cell+plasticity+in+a+human+organotypic+melanoma+culture.">The tumour microenvironment shapes dendritic cell plasticity in a human organotypic melanoma culture.</a> Nature Communications. 11 (1), 2749 (2020).</li><li>van Beek, J. J. P., et al. <a target="_blank" href="http://www.ncbi.nlm.nih.gov/entrez/query.fcgi?db=PubMed&cmd=Search&doptcmdl=Citation&defaultField=Title+Word&term=Human+pDCs+are+superior+to+cDC2s+in+attracting+cytolytic+lymphocytes+in+melanoma+patients+receiving+DC+vaccination.">Human pDCs are superior to cDC2s in attracting cytolytic lymphocytes in melanoma patients receiving DC vaccination.</a> Cell Reports. 30 (4), 1027-1038 (2020).</li><li>Rodriguez-Rosales, Y. A., et al. <a target="_blank" href="http://www.ncbi.nlm.nih.gov/entrez/query.fcgi?db=PubMed&cmd=Search&doptcmdl=Citation&defaultField=Title+Word&term=Immunomodulatory+aged+neutrophils+are+augmented+in+blood+and+skin+of+psoriasis+patients.">Immunomodulatory aged neutrophils are augmented in blood and skin of psoriasis patients.</a> Journal of Allergy and Clinical Immunology. 148 (4), 1030-1040 (2021).</li><li>Hoeijmakers, Y. M., et al. <a target="_blank" href="http://www.ncbi.nlm.nih.gov/entrez/query.fcgi?db=PubMed&cmd=Search&doptcmdl=Citation&defaultField=Title+Word&term=Immune+cell+composition+in+the+endometrium+of+patients+with+a+complete+molar+pregnancy:+Effects+on+outcome.">Immune cell composition in the endometrium of patients with a complete molar pregnancy: Effects on outcome.</a> Gynecologic Oncology. 160 (2), 450-456 (2021).</li><li>van Ravensteijn, S. G., et al. <a target="_blank" href="http://www.ncbi.nlm.nih.gov/entrez/query.fcgi?db=PubMed&cmd=Search&doptcmdl=Citation&defaultField=Title+Word&term=Immunological+and+genomic+analysis+reveals+clinically+relevant+distinctions+between+angiosarcoma+subgroups.">Immunological and genomic analysis reveals clinically relevant distinctions between angiosarcoma subgroups.</a> Cancers. 14 (23), 5938 (2022).</li><li>van der Woude, L. L., et al. <a target="_blank" href="http://www.ncbi.nlm.nih.gov/entrez/query.fcgi?db=PubMed&cmd=Search&doptcmdl=Citation&defaultField=Title+Word&term=Tumor+microenvironment+shows+an+immunological+abscopal+effect+in+patients+with+NSCLC+treated+with+pembrolizumab-radiotherapy+combination.">Tumor microenvironment shows an immunological abscopal effect in patients with NSCLC treated with pembrolizumab-radiotherapy combination.</a> Journal for Immunotherapy of Cancer. 10 (10), e005248 (2022).</li><li>Graham Martinez, C., et al. <a target="_blank" href="http://www.ncbi.nlm.nih.gov/entrez/query.fcgi?db=PubMed&cmd=Search&doptcmdl=Citation&defaultField=Title+Word&term=The+immune+microenvironment+landscape+shows+treatment-specific+differences+in+rectal+cancer+patients.">The immune microenvironment landscape shows treatment-specific differences in rectal cancer patients.</a> Frontiers in Immunology. 13, 1011498 (2022).</li><li>Cortenbach, K. R. G., et al. <a target="_blank" href="http://www.ncbi.nlm.nih.gov/entrez/query.fcgi?db=PubMed&cmd=Search&doptcmdl=Citation&defaultField=Title+Word&term=Topography+of+immune+cell+infiltration+in+different+stages+of+coronary+atherosclerosis+revealed+by+multiplex+immunohistochemistry.">Topography of immune cell infiltration in different stages of coronary atherosclerosis revealed by multiplex immunohistochemistry.</a> International Journal of Cardiology. Heart &amp; Vasculature. 44, 101111 (2023).</li><li>van Wilpe, S., et al. <a target="_blank" href="http://www.ncbi.nlm.nih.gov/entrez/query.fcgi?db=PubMed&cmd=Search&doptcmdl=Citation&defaultField=Title+Word&term=Homologous+recombination+repair+deficient+prostate+cancer+represents+an+immunologically+distinct+subtype.">Homologous recombination repair deficient prostate cancer represents an immunologically distinct subtype.</a> Oncoimmunology. 11 (1), 2094133 (2022).</li><li>Gorris, M. A. J., et al. <a target="_blank" href="http://www.ncbi.nlm.nih.gov/entrez/query.fcgi?db=PubMed&cmd=Search&doptcmdl=Citation&defaultField=Title+Word&term=Paired+primary+and+metastatic+lesions+of+patients+with+ipilimumab-treated+melanoma:+high+variation+in+lymphocyte+infiltration+and+HLA-ABC+expression+whereas+tumor+mutational+load+is+similar+and+correlates+with+clinical+outcome.">Paired primary and metastatic lesions of patients with ipilimumab-treated melanoma: high variation in lymphocyte infiltration and HLA-ABC expression whereas tumor mutational load is similar and correlates with clinical outcome.</a> Journal for Immunotherapy of Cancer. 10 (5), e004329 (2022).</li><li>van Wilpe, S., et al. <a target="_blank" href="http://www.ncbi.nlm.nih.gov/entrez/query.fcgi?db=PubMed&cmd=Search&doptcmdl=Citation&defaultField=Title+Word&term=Intratumoral+T+cell+depletion+following+neoadjuvant+chemotherapy+in+patients+with+muscle-invasive+bladder+cancer+is+associated+with+poor+clinical+outcome.">Intratumoral T cell depletion following neoadjuvant chemotherapy in patients with muscle-invasive bladder cancer is associated with poor clinical outcome.</a> Cancer Immunology, Immunotherapy. 72 (1), 137-149 (2023).</li><li>Sultan, S., Gorris, M. A. J., Buytenhuijs, F., Lvan de Woude, L. <a target="_blank" href="http://www.ncbi.nlm.nih.gov/entrez/query.fcgi?db=PubMed&cmd=Search&doptcmdl=Citation&defaultField=Title+Word&term=A+Segmentation-free+machine+learning+architecture+for+immune+land-scape+phenotyping+in+solid+tumors+by+multichannel+imaging.">A Segmentation-free machine learning architecture for immune land-scape phenotyping in solid tumors by multichannel imaging.</a> bioRxiv. , (2021).</li><li>Parra, E. R., et al. <a target="_blank" href="http://www.ncbi.nlm.nih.gov/entrez/query.fcgi?db=PubMed&cmd=Search&doptcmdl=Citation&defaultField=Title+Word&term=Immuno-profiling+and+cellular+spatial+analysis+using+five+immune+oncology+multiplex+immunofluorescence+panels+for+paraffin+tumor+tissue.">Immuno-profiling and cellular spatial analysis using five immune oncology multiplex immunofluorescence panels for paraffin tumor tissue.</a> Scientific Reports. 11 (1), 8511 (2021).</li><li>Parra, E. R., et al. <a target="_blank" href="http://www.ncbi.nlm.nih.gov/entrez/query.fcgi?db=PubMed&cmd=Search&doptcmdl=Citation&defaultField=Title+Word&term=Validation+of+multiplex+immunofluorescence+panels+using+multispectral+microscopy+for+immune-profiling+of+formalin-fixed+and+paraffin-embedded+human+tumor+tissues.">Validation of multiplex immunofluorescence panels using multispectral microscopy for immune-profiling of formalin-fixed and paraffin-embedded human tumor tissues.</a> Scientific Reports. 7 (1), 13380 (2017).</li><li>Sun, Z., Nyberg, R., Wu, Y., Bernard, B., Redmond, W. L. <a target="_blank" href="http://www.ncbi.nlm.nih.gov/entrez/query.fcgi?db=PubMed&cmd=Search&doptcmdl=Citation&defaultField=Title+Word&term=Developing+an+enhanced+7-color+multiplex+IHC+protocol+to+dissect+immune+infiltration+in+human+cancers.">Developing an enhanced 7-color multiplex IHC protocol to dissect immune infiltration in human cancers.</a> PLoS One. 16 (2), e0247238 (2021).</li><li>LeicaBiosystems. <a target="_blank" href="http://www.ncbi.nlm.nih.gov/entrez/query.fcgi?db=PubMed&cmd=Search&doptcmdl=Citation&defaultField=Title+Word&term=BOND+RX+Fully+Automated+Research+Stainer+Protocols.">BOND RX Fully Automated Research Stainer Protocols.</a> , (2022).</li></ol>

The authors have no conflicts of interest to disclose.

Spatial analysis of the TME is a sought-after technique to learn more about the immune cell compartment and discover new prognostic and predictive biomarkers, particularly in the field of immuno-oncology16. Many different techniques are being developed for this purpose, involving the detection of proteins, mRNA transcripts, or a combination of the two, with estimations up to 100-1,000 targets. However, higher multiplexing comes at the cost of less high-throughput experiments, higher experimental costs, and technical challenges, and often, only a small part of the TME can be analyzed. Multiplex IHC using the TSA-based method that we describe here, detects six different markers + DAPI simultaneously, is relatively less expensive to perform, and whole tissue sections are imaged in under 20 min, ready to be analyzed fully. This technique has become less complex with the automation of the staining procedure. Improvements in the multispectral microscope, which include the addition of two extra filters, have improved spectral unmixing and scanning times tremendously. It is possible to detect up to eight different markers + DAPI simultaneously. However, by expanding the multiplexing with more markers, the aforementioned benefits disappear as spectral unmixing becomes more challenging and scanning times for whole slides increase substantially. Efforts are being undertaken to standardize multiplex IHC between different institutions to facilitate implementation in the diagnostic setting more easily. For this standardization of multiplex IHC, we advise users to adhere to the more accessible protocol with six different markers + DAPI. Nevertheless, still quite some technical know-how is necessary and downstream analysis can be challenging, for which we have developed methodologies that are described in this protocol.
Standardization begins with multiplex IHC panel development. The importance of the choice of primary antibodies detecting particular protein targets has been emphasized before17. Our multiplex IHC panels are mostly developed with primary antibody clones that are also used and validated for IHC at our diagnostics department. However, in the case of the dendritic cell multiplex IHC panel, most antibodies were not used in the diagnostic setting (van der Hoorn et al., manuscript in submission). To ensure specificity and minimize batch differences, we chose to use monoclonal antibodies over polyclonal antibodies and have also validated most antibodies using transfected cell lines and primary cells. Over the years, different versions of multiplex IHC panels have been used in numerous studies using the Vectra 3 system18,21,23,24,25,26,27,28,29,30,31,32. To implement these multiplex IHC panels optimally on the PhenoImager HT system, some adjustments had to be made in primary antibody and fluorophore combinations. To benefit from better spectral unmixing and faster scanning times of whole tissue sections, implementation of the latest Opal480 and Opal780 fluorophores and avoiding the use of Opal540 and Opal650 fluorophores in seven-color multiplex IHC panels is necessary. Scanning times are ~3-10 times faster depending on the size of the tissue section. Multiplex IHC panel adjustments were quite easy to achieve, but some considerations need to be kept in mind. The fluorescent spectrum of Opal480 overlaps a lot with the autofluorescence spectrum and therefore, interferes with the spectral unmixing of erythrocytes and other autofluorescent structures. Using an increased concentration of the primary antibody paired with Opal480 solved this issue in most cases. The implementation of the proprietary Sample AF filter on the PhenoImager HT facilitates the unmixing of Opal480 and autofluorescence. However, it is best to use a primary antibody that yields a clear signal when used with Opal480 so that its signal is higher than the autofluorescence.
Even though these multiplex IHC panels are established, batch-to-batch variation is something that needs to be considered. By performing monoplex IHC controls before starting the full multiplex IHC experiment, we sometimes observed that primary antibodies perform either stronger or weaker from experiment to experiment. The reasons for this could be pipetting errors, suboptimal reagent storage conditions, and shelf life. We solved this by adjusting the primary antibody solution based on our experience. Even when none of the aforementioned adjustments had to be made, with every multiplex IHC batch experiment, it is important to set exposure times based on monoplex IHC-stained control slides.
Because our research was initially focused on different types of carcinomas and melanoma, multiplex IHC panels were required to be interchangeable between tumor types with minimal adjustments. Therefore, we always included multiple (tumor) tissue types in the optimization process and observed that dilutions for primary antibodies for immune cell markers can be kept similar between different tumor types. However, tumor tissue detection between carcinomas and melanoma needs different tumor markers. Accordingly, the tumor marker was always optimized to work at the end of each multiplex IHC panel and is currently always used in conjunction with Opal780, which coincidentally also has to be at the last fluorophore in a multiplex IHC staining procedure. By using the tumor marker consequently at the end of the multiplex IHC, these multiplex IHC panels can be easily exchanged for other tumor types, such as glioblastoma (i.e., GFAP) and Hodgkin lymphoma (i.e., CD30). For angiosarcoma, we used this lymphocyte multiplex IHC panel with erythroblast transformation-specific-related gene (ERG) as the tumor marker with only two optimization experiments25. The optimization included titration of the ERG primary antibody and testing the multiplex IHC panel with ERG at the end.
Other adjustments to these multiplex IHC panels can also be made by exchanging a certain immune cell marker for another immune or functional marker. Every change requires optimization. The protocol for optimization could be followed as described previously17. Certain changes to the proposed multiplex IHC panels will interfere with the ImmuNet algorithms that we have created. Sufficient data must be generated and time has to be spent to implement these changes into the algorithm (at least 750 annotations for every new marker and/or cell phenotypes, and 150 annotations for validation of previously trained markers). The panels presented here do not contain functional markers, although the implementation of immune checkpoint markers such as PD-1 and PD-L1 into multiplex IHC panels is performed in our laboratory. However, the analysis of markers that are less binary in negative and positive signals has proven to be more difficult and is an area of active research in our group.
The number of markers that can be simultaneously assessed with multiplex IHC is limited compared to other novel techniques. While this can be circumvented by analyzing different panels on consecutive slices of an FFPE block, it will be hard to compare these slices spatially. Orientation and folded artifacts are likely not the same after slide preparation. Nevertheless, multiplex IHC is quite accessible, which makes it an attractive tool for more institutions and researchers and therefore, more suitable for future implementation in a diagnostic setting. With the standardization of multiplex IHC immune cell panels for multiple tumor types and downstream analysis pipelines, more knowledge could be gained about differences in TME between patients and tumor types. This can, for instance, lead to more insights into the role of the TME in antitumor response to specific treatments. This may even give rise to new biomarkers to predict factors such as response to treatment and expected survival. Overall, this can enable multiplex IHC to become a clinical tool to aid with clinical decision-making, in a personalized medicine approach. Admittedly, more steps of the analysis procedure should probably be automated and standardized for it to be feasible for use in a daily diagnostic setting, so as of yet, it is mostly a futuristic perspective.
Analysis of multiple markers on a single sample slide can be a very powerful tool in spite of its technical challenges. With standardized experimental protocols and a robust analysis method, as we described here using ImmuNet, the quantification of multiple markers makes it more informative than classical IHC, while multiplex IHC remains relatively high-throughput compared to novel higher plex experimental methods.

Immune cells play a crucial role in the protection against pathogens such as viruses and bacteria, but also against cancerous cells1. Therefore, the immune system within the tumor microenvironment (TME) holds a lot of promise for discovering prognostic and predictive biomarkers2. Immune cell infiltrates have been correlated to prognosis in various types of cancer, although this has not been implemented in clinical care yet3,4. In most tumor types, high numbers of cytotoxic T cells and T helper 1 cells and/or low numbers of regulatory T cells are linked to good prognoses. Efforts are ongoing to incorporate a so-called &#34;Immunoscore&#34; into the TNM staging of colorectal cancer, turning it into TNM-I staging5,6. The Immunoscore is derived from the total number of T cells (detected with CD3) and cytotoxic T cells (detected with CD8) in two different tumor regions: the tumor core versus the invasive margin (IM) of tumors. The Immunoscore has also been proposed to be of prognostic value in other cancer types, such as melanoma, lung cancer, and breast cancer6,7,8,9. Furthermore, immune cell infiltrates may also correlate to response to checkpoint blockade immunotherapy10. However, these predictive biomarkers must be validated in prospective studies before they can be routinely implemented in clinical practice. Moreover, it has also been proposed that a single biomarker will be insufficient for meaningful prediction11. Therefore, creating a complete map of a patient sample by combining different biomarkers has been proposed as a more comprehensive predictive biomarker in a so-called &#34;cancer immunogram&#34;12.
Among the methods for studying immune cells within the TME, the oldest and most well-known technique is immunohistochemistry (IHC), routinely used for diagnostic testing in several diseases, especially cancer13. This technique was limited to the use of one or only a few markers14 for a long time and therefore, was outcompeted in research settings by other techniques such as flow cytometry and gene expression profiling (GEP). However, the formalin-fixed and paraffin-embedded (FFPE) tumor tissues typically used in routine diagnostics and research are not (optimally) suitable for flow cytometry and GEP. Furthermore, although GEP and flow cytometry provide a lot of insight into cell phenotype and function, the lack of spatial information is a major disadvantage. Therefore, heterogeneity within a sample, such as differences in immune cell-infiltrated versus immune cell-excluded areas of a tumor, could go undetected15. Novel platforms have been developed for multiplex analysis of FFPE tissues, such as multiplex IHC, imaging mass cytometry, and CO-Detection by indEXing (CODEX) that can be used to detect multiple markers simultaneously within a tissue section16. Immune cells in the TME are being widely studied to find the best biomarkers for immunotherapy. However, multiplex techniques and automated image analysis pose hurdles of their own.
Our laboratory has extensive experience in multiplex IHC staining using the Opal/Tyramide signal amplification (TSA) method and has automated this on a&#160;IHC platform (see the Table of Materials)17,18,19,20,21,22,23,24,25,26,27,28,29,30,31. We have optimized immune cell panels for the detection of different subsets of lymphocytes, myeloid cells, and dendritic cells (DCs). Tissues that contain dense immune cell areas - for lymphocytes or complex cell morphologies (i.e., myeloid cells and DCs) - are particularly challenging to analyze, with a risk of over- or underestimating the number of immune cells present. To overcome this problem, ImmuNet analysis software was developed by our group32, and this machine-learning pipeline improved the quality of the detection of these different types of immune cells immensely. A detailed protocol from obtaining the FFPE material to the analysis of immune cell densities in different tissue compartments and distances between immune cell types is described here.
This protocol outlines how the multiplex IHC panels are performed at the Radboud University Medical Center since the implementation of the digital pathology imager in 2022. The described multiplex IHC panels can be used for different carcinomas (e.g., lung, prostate, colorectal, bladder, breast) with the use of a pan-cytokeratin antibody as a tumor marker or for melanoma with the use of melanocyte-associated antibodies as tumor markers. These multiplex IHC protocols have been carefully optimized in terms of primary antibody concentration, fluorophore combinations, and the sequence of the staining procedure. We and others have described multiplex IHC panel optimization earlier17,33,34,35. Multiplex IHC panels can be adapted, but the described analysis pipelines need to be evaluated and potentially adjusted or retrained accordingly. The described seven-color multiplex IHC protocols make use of the Opal fluorophores Opal480, Opal520, Opal570, Opal620, Opal690, Opal780, and 4&#39;,6-diamidino-2-phenylindole (DAPI), so that easy unmixing and fast scanning on the imager is enabled with &#34;Multispectral One Touch ImmunoFluorescence&#34; (MOTiF). Nine-color staining and scanning is not described in this protocol as this requires even more finetuning of the experimental setup and another mode of scanning on the imager that uses the liquid crystal tunable filter.

<table border="1" fo:keep-together.within-page="1" fo:keep-with-next.within-page="always">
	<tbody>
		<tr>
			<td>anti-CD14</td>
			<td>Cell Marque&#160;</td>
			<td>114R-16</td>
			<td>section 3, clone EPR3653</td>
		</tr>
		<tr>
			<td>anti-CD163</td>
			<td>Cell Marque&#160;</td>
			<td>163M-15</td>
			<td>section 3, clone MRQ-26</td>
		</tr>
		<tr>
			<td>anti-CD19</td>
			<td>Abcam</td>
			<td>ab134114</td>
			<td>section 3, clone EPR5906</td>
		</tr>
		<tr>
			<td>anti-CD1c (BDCA1)</td>
			<td>Thermo Scientific</td>
			<td>TA505411</td>
			<td>section 3, clone OTI2F4</td>
		</tr>
		<tr>
			<td>anti-CD20</td>
			<td>Thermo Scientific</td>
			<td>MS-340-S</td>
			<td>section 3, clone L26</td>
		</tr>
		<tr>
			<td>anti-CD3</td>
			<td>Thermo Scientific</td>
			<td>RM-9107</td>
			<td>section 3, clone sp7</td>
		</tr>
		<tr>
			<td>anti-CD303/BDCA2</td>
			<td>Dendritics via Enzo Lifesciences/Axxora</td>
			<td>DDX0043</td>
			<td>section 3, clone 124B3.13</td>
		</tr>
		<tr>
			<td>anti-CD56</td>
			<td>Cell Marque&#160;</td>
			<td>156R-94</td>
			<td>section 3, clone MRQ-42</td>
		</tr>
		<tr>
			<td>anti-CD66b</td>
			<td>BD Biosciences</td>
			<td>555723</td>
			<td>section 3, clone G10F5</td>
		</tr>
		<tr>
			<td>anti-CD68</td>
			<td>Dako Agilent</td>
			<td>M087601</td>
			<td>section 3, clone PG-M1</td>
		</tr>
		<tr>
			<td>anti-CD8</td>
			<td>Dako Agilent</td>
			<td>M7103</td>
			<td>section 3, clone C8/144B</td>
		</tr>
		<tr>
			<td>anti-Foxp3</td>
			<td>Thermo Scientific</td>
			<td>14-4777</td>
			<td>section 3, clone 236A/E7</td>
		</tr>
		<tr>
			<td>anti-Gp100</td>
			<td>Dako Agilent</td>
			<td>M063401</td>
			<td>section 3, clone HMB45</td>
		</tr>
		<tr>
			<td>anti-HLA-DR, DP, DQ</td>
			<td>Santa Cruz</td>
			<td>sc-53302</td>
			<td>section 3, clone CR3/43</td>
		</tr>
		<tr>
			<td>anti-MART-1</td>
			<td>Thermo Scientific</td>
			<td>MS-799</td>
			<td>section 3, clone A103</td>
		</tr>
		<tr>
			<td>anti-pan cytokeratin</td>
			<td>Abcam</td>
			<td>ab86734</td>
			<td>section 3, clone AE1/AE3 + 5D3</td>
		</tr>
		<tr>
			<td>anti-SOX10</td>
			<td>Sigma Aldrich</td>
			<td>383R</td>
			<td>section 3, clone EP268</td>
		</tr>
		<tr>
			<td>anti-Tyrosinase</td>
			<td>Sanbio</td>
			<td>MONX10591</td>
			<td>section 3, clone T311</td>
		</tr>
		<tr>
			<td>anti-XCR1</td>
			<td>Cell Signaling Technologies via Biok&#233;</td>
			<td>44665S</td>
			<td>section 3, clone D2F8T</td>
		</tr>
		<tr>
			<td>antibody diluent</td>
			<td>Akoya BioSciences</td>
			<td>SKU&#160;ARD1001EA</td>
			<td>section 3, from Opal 7-Color Automation IHC Kit 50 slide (can optionally also be replaced by TBST with 10% BSA)</td>
		</tr>
		<tr>
			<td>Bond Aspirating Probe</td>
			<td>Leica Biosciences</td>
			<td>S21.0605</td>
			<td>section 3</td>
		</tr>
		<tr>
			<td>Bond Aspirating Probe Cleaning</td>
			<td>Leica Biosciences</td>
			<td>CS9100</td>
			<td>section 3</td>
		</tr>
		<tr>
			<td>Bond Dewax Solution</td>
			<td>Leica Biosciences</td>
			<td>AR9222</td>
			<td>section 3</td>
		</tr>
		<tr>
			<td>Bond Objectglas label + print lint</td>
			<td>Leica Biosciences</td>
			<td>S21.4564.A</td>
			<td>section 3</td>
		</tr>
		<tr>
			<td>Bond Research Detection System 2</td>
			<td>Leica Biosciences</td>
			<td>DS9777</td>
			<td>section 3</td>
		</tr>
		<tr>
			<td>Bond RX autostainer&#160;</td>
			<td>Leica Biosciences</td>
			<td>-</td>
			<td>section 3, automated platform</td>
		</tr>
		<tr>
			<td>Bond TM Epitope Retrieval 1 - 1 L</td>
			<td>Leica Biosciences</td>
			<td>AR9961</td>
			<td>section 3</td>
		</tr>
		<tr>
			<td>Bond TM Epitope Retrieval 2 - 1 L</td>
			<td>Leica Biosciences</td>
			<td>AR9640</td>
			<td>section 3</td>
		</tr>
		<tr>
			<td>Bond TM Wash Solution 10x - 1 L</td>
			<td>Leica Biosciences</td>
			<td>AR9590</td>
			<td>section 3</td>
		</tr>
		<tr>
			<td>BOND Universal Covertile</td>
			<td>Leica Biosciences</td>
			<td>S21.4611</td>
			<td>section 3</td>
		</tr>
		<tr>
			<td>Bond(TM) Titration Kit</td>
			<td>Leica Biosciences</td>
			<td>OPT9049</td>
			<td>section 3</td>
		</tr>
		<tr>
			<td>Coverslip 24 x 32 mm #1 (0.13-0.16 mm)</td>
			<td>Fisher Scientific</td>
			<td>15717592</td>
			<td>section 2</td>
		</tr>
		<tr>
			<td>coverslip 24 x 50 mm</td>
			<td>VWR</td>
			<td>631-0146</td>
			<td>section 2</td>
		</tr>
		<tr>
			<td>DAPI Fluoromount-G</td>
			<td>VWR</td>
			<td>0100-20</td>
			<td>section 3, whenever monoplex slides need to be quickly checked, not for official analysis, then DAPI is stained seperately for better results</td>
		</tr>
		<tr>
			<td>Eosine</td>
			<td></td>
			<td></td>
			<td>section 2, home made</td>
		</tr>
		<tr>
			<td>Ethanol 99.5%</td>
			<td>VWR</td>
			<td>4099.9005</td>
			<td>section 2</td>
		</tr>
		<tr>
			<td>Fluoromount-G</td>
			<td>VWR</td>
			<td>0100-01</td>
			<td>section 3&#160;</td>
		</tr>
		<tr>
			<td>haematoxyline</td>
			<td></td>
			<td>-</td>
			<td>section 2, home made</td>
		</tr>
		<tr>
			<td>ImmuNet</td>
			<td></td>
			<td>-</td>
			<td>immune cell detection and phenotyping pipeline</td>
		</tr>
		<tr>
			<td>inForm software 2.4.10</td>
			<td>Akoya BioSciences</td>
			<td>-</td>
			<td>section 4 &#38; 6</td>
		</tr>
		<tr>
			<td>OPAL 480 reagent pack</td>
			<td>Akoya BioSciences</td>
			<td>FP1500001KT</td>
			<td>section 3</td>
		</tr>
		<tr>
			<td>OPAL 520 reagent pack</td>
			<td>Akoya BioSciences</td>
			<td>FP1487001KT</td>
			<td>section 3</td>
		</tr>
		<tr>
			<td>OPAL 570 reagent pack</td>
			<td>Akoya BioSciences</td>
			<td>FP1488001KT</td>
			<td>section 3</td>
		</tr>
		<tr>
			<td>OPAL 620 reagent pack</td>
			<td>Akoya BioSciences</td>
			<td>FP1495001KT</td>
			<td>section 3</td>
		</tr>
		<tr>
			<td>OPAL 690 reagent pack</td>
			<td>Akoya BioSciences</td>
			<td>FP1497001KT</td>
			<td>section 3</td>
		</tr>
		<tr>
			<td>OPAL 780 reagent pack</td>
			<td>Akoya BioSciences</td>
			<td>FP1501001KT</td>
			<td>section 3</td>
		</tr>
		<tr>
			<td>Opal 7-Color Automation IHC Kit 50 slide</td>
			<td>Akoya BioSciences</td>
			<td>NEL821001KT</td>
			<td>section 3</td>
		</tr>
		<tr>
			<td>PhenoChart 1.1.0</td>
			<td>Akoya BioSciences</td>
			<td>-</td>
			<td>section 5</td>
		</tr>
		<tr>
			<td>PhenoImagerHT</td>
			<td>Akoya BioSciences</td>
			<td>CLS143455</td>
			<td>section 4, digital pathology imager with slide viewer and imaging software (formerly known as Vectra Polaris)</td>
		</tr>
		<tr>
			<td>Quick-D mounting medium</td>
			<td>Klinipath</td>
			<td>7280</td>
			<td>section 2</td>
		</tr>
		<tr>
			<td>QuPath 0.3.2</td>
			<td></td>
			<td></td>
			<td>whole slide image analysis software platform</td>
		</tr>
		<tr>
			<td>R 4.1.1</td>
			<td></td>
			<td></td>
			<td></td>
		</tr>
		<tr>
			<td>Slide boxes</td>
			<td>VWR</td>
			<td>631-0737</td>
			<td>section 1</td>
		</tr>
		<tr>
			<td>SuperFrost Plus</td>
			<td>Thermo Scientific through VWR</td>
			<td>631-9483</td>
			<td>section 1</td>
		</tr>
		<tr>
			<td>Vectra Polaris software 1.0.13</td>
			<td>Akoya BioSciences</td>
			<td>-</td>
			<td>section 4</td>
		</tr>
		<tr>
			<td>Xylene</td>
			<td>VWR</td>
			<td>4055-9005</td>
			<td>section 2</td>
		</tr>
	</tbody>
</table>

multiplex immunohistochemical analysis of the spatial immune cell landscape of the tumor microenvironment

The immune cell landscape of the tumor microenvironment potentially contains information for the discovery of prognostic and predictive biomarkers. Multiplex immunohistochemistry is a valuable tool to visualize and identify different types of immune cells in tumor tissues while retaining its spatial information. Here we provide detailed protocols to analyze lymphocyte, myeloid, and dendritic cell populations in tissue sections. Starting from cutting formalin-fixed paraffin-embedded sections, automatic multiplex staining procedures on an automated platform, scanning of the slides on a multispectral imaging microscope, to the analysis of images using an in-house-developed machine learning algorithm ImmuNet. These protocols can be applied to a variety of tumor specimens by simply switching tumor markers to analyze immune cells in different compartments of the sample (tumor versus invasive margin) and apply nearest-neighbor analysis. This analysis is not limited to tumor samples but can also be applied to other (non-)pathogenic tissues. Improvements to the equipment and workflow over the past few years have significantly shortened throughput times, which facilitates the future application of this procedure in the diagnostic setting.

Author Spotlight: Unlocking Insights into the Immune Cell Landscape of Tumors

Multiplex Immunohistochemical Analysis of the Spatial Immune Cell Landscape of the Tumor Microenvironment

Our aim is to identify the amount, type, and location of immune cells within the tumor microenvironments. In the field of immunochemistry, there have been major advance in technology in recent years. Advance in multiplex immunochemistry technology allow for more markers to be reliably studied.Current experimental challenges are visualize specific types of immune cells within tumor samples, which have been overcome by using the multiplex immunochemistry methods. However, accurate automatic analysis of these images is still a challenge that we're working on. The differences between immune cell landscape between primary and metastatic tumors generate questions such as, what tumor intrinsic properties drive these differences, and what do these differences mean for therapies that target the immune system?Next to our exploratory work, we will also focus on clinically relevant questions such as how the immune cell landscape of the tumor is affected by different types of anti-cancer therapies and whether the immune cell landscape could predict the outcome of therapy.

Read this Scientific Article Protocol about Multiplex Immunohistochemical Analysis of the Spatial Immune Cell Landscape of the Tumor Microenvironment at JoVE.com

This protocol describes in detail how immune cell characterization of the tumor microenvironment using multiplex immunohistochemistry is carried out.

The immune cell landscape of the tumor microenvironment potentially contains information for the discovery of prognostic and predictive biomarkers. ...

multiplex-immunohistochemical-analysis-spatial-immune-cell-landscape

Research

JoVE Journal

Cancer Research

Workflow for Immune Cells Detection Using the Machine Learning Algorithm ImmuNet

After performing immunohistochemistry on the FFPE tissues, image the sections using the digital pathology imager. Launch the software to perform spectral unmixing of the annotated images. Select File, then click Open Image, and choose the QPTIFF files to upload the images into the software.Allow the stamps marked as informed projects to upload into the project. Load the QPTIFF files prepared for the autofluorescence compensation. Select the autofluorescence on the image tool to draw a line on the image from the unstained slide through autofluorescent structures like erythrocytes and collagen to compensate for autofluorescence.Then navigate to the Edit Markers and Colors section. Assign marker names that match the Opal fluorophore, and adjust the color settings to your preferred option. Select Prepare All located in the lower left corner of the interface to initiate the unmixing of fluorophores.Examine all images to check the visibility of all signals and the successful completion of the unmixing process. Select the eyeball icon to turn off and on all the markers one by one to check the quality. Now navigate to the Export tab and create a new empty export directory by clicking the Browse button located under the Export directory.Select composite image and component images multi-image TIFF under the images to export tab. Navigate to the Batch Analysis tab found vertically on the left for batch processing of slides and select create separate directories for each item under the export options. To add slides for analysis, select QPTIFF files under the Add Slides button and load these into the batch analysis.Select Run to commence the batch processing of slides. Create a new folder containing only the component files from spectral unmixing, ensuring the hierarchical folder structure remains intact. Launch the whole slide viewer software.Click Create Project on the left side and create or select a new empty folder with an appropriate name. Then click Automate and choose Show script editor. Copy and paste the existing script and modify the location to direct it toward the folder that contains all the slide component files.Select Run to begin the process of batch stitching slides, and then allow it to complete. Now move the recently generated OME TIFF files into the QuPath project and save them as a new project. When a new window appears, select set image type as fluorescence, then click the Import button.Navigate to the menu on the left, select a sample from the list and double click to open the chosen sample. Adjust the intensity of the channels by clicking the Contrast icon to enhance visibility. Select all channels and opt to reset.Ensure the autofluorescence is toggled off. To draw a region of interest or ROI for the tumor, click the contrast icon again then select Show grayscale. After that, select the tumor marker channel and adjust the intensity for optimal visibility.Click the brush tool to draw an ROI around the tumor. Click the one tool and adjust the ROI while pressing the alt key to smooth the ROI from the outside. When finished, merge any separated parts of the tumor into the same ROI.Right click the ROI annotation in the list on the left, select Set Properties, and provide a suitable name. For instance, Tumor. Expand the ROI for the invasive margin from the tumor region by selecting Objects, then Annotations, followed by Expand Annotations.Choose the desired size for the expansion radius. Select remove interior and constrain to parent. Click on the contrast icon, select the autofluorescence channel and adjust the intensity for optimal visibility.Click the wand and adjust the ROI while pressing the alt key to smooth the ROI from the outside and remove any background that should not be part of this ROI. Right click the annotation in the list on the left. Then select Set Properties to assign a suitable name, such as invasive margin or IM to the ROI.If desired, change its color. To export the annotations, navigate to the File option, click Object Data, Export as GeoJSON, and select Export All Objects. Continue with the default selection on export as feature collection, and save it in a preferred location.

1.3K Views.  Radboudumc.  This protocol describes in detail how immune cell characterization of the tumor microenvironment using multiplex immunohistochemistry is carried out.

Mapping the Tumor Immune Landscape: An In-Depth Multiplex IHC Analysis Protocol

Summary